Neurofibromatosis type 1 (NF1) is a common inherited disease syndrome caused by germline mutations in the NF1 gene. About one third of NF1 patients develop diffuse, plexiform neurofibromas that can transform to malignant peripheral nerve sheath tumors - a cancer that is frequently fatal. Remarkably, in human tumors and in mouse models of NF1, neurofibromas almost invariably contain NF1-null Schwann cells and NF1-heterozygous mast cells. Transplanting such NF1-prone mice with wild-type bone marrow prevents tumorigenesis, implying that heterozygous bone marrow-derived cells such as mast cells are a required component in pathogenesis, and that targeting signaling pathways in either Schwann cells or in mast cells might be of therapeutic benefit. The NF1 gene encodes a large protein, termed Neurofibromin, with GTPase Activating Protein (GAP) activity towards Ras. Complete or hemizygous loss of the NF1 gene leads to increased K-ras activity in both Schwann cells and mast cells, with concomitant activation of downstream effectors that promote proliferation and changes in cell shape and movement. Recently, we have shown that p21-activated kinases (Paks) play a vital role in K-ras signaling, in particular in the activation of te Erk and Akt pathways. As clinical-grade small molecules Pak inhibitors have recently emerged, there is a direct line from our proposed experiments to therapeutic application in NF1 syndrome, and perhaps also in other cancers driven by acquired, somatic NF1 mutations. In addition, we have used a new phospho-proteomic method to analyze the activity of hundreds of protein kinases in NF1-deficient cells, confirming known K-ras-activated signaling activity (e.g., elevated activity of Pak, Erk, and Akt/mTOR pathways) but also uncovering new potential candidates for targeted therapy. In this proposal, we postulate that Paks, which are required both for Erk and Akt/mTOR pathway activation downstream of K-ras, are uniquely suited as targets for therapy in NF1-driven plexiform neurofibromas and malignant peripheral nerve sheath tumors. Further, we posit that, should resistance emerge to anti-Pak agents, whole kinome analysis can be used to identify likely escape routes, allowing for more effective combinatorial therapy. Accordingly, we propose three aims: 1) We will determine the baseline status of the kinome in NF1-/- Schwann cells and mast cells, and how the kinome reprograms under pressure of Mek, Akt/mTOR, or Pak inhibition; 2) We will use mouse models of NF1 and Pak to determine the cellular basis for Pak's function in NF1-related tumors; and 3) We will determine the efficacy of specific small molecule Pak inhibitors in NF1 mouse models and xenografts, and assess the signaling responses elicited by such agents in vivo.